When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
timing belt job causes rich AFR? What have I broken this time?
Wall of text…
So I got my car tuned for turbo, and life was good. But the slow oil leak at the front of the motor had become a pretty messy oil leak, so I decided to replace the cam & crank seals. As I was putting it all back together, I realized that I hadn’t followed the correct belt tensioning procedure the last time I did a timing belt job about a year ago. I was surprised how much belt slack the tensioner sucks up when you rotate the crank 1&5/6 rotations & use the tensioning mark. I’m surprised I didn’t have any issues with the belt previously. Here’s where things started to go south: I was being lazy when I torqued the crank bolt back on, and instead of using the FM tool to brace against the water pump (WHICH I OWN) I was reacting the crank bolt torque with a BFwrench on the intake cam. Before I hit final torque, something metallic went “crack!”, which is always a bad sign…
I wasn’t sure if it was my torque wrench or some unknown car part that had made the noise, but I buttoned everything back up anyway. Guess what? It ran like total *** when I tried to start it. Timing was surging up into the high 20’s, and AFR’s were so nuts, it temporarily killed my EGO sensor such that it read max lean (22+) for a while before trickling back down to a sane reading (more to come on this). So I thought maybe that crack noise I had heard while torqueing was something horrible happening to the VVT actuator on the intake cam. I have a spare VVT head, so I redid the timing belt and swapped in the other VVT. When I redid the timing belt, I noticed that the crank sensor was basically resting against the timing wheel teeth- really small gap. Maybe this was my issue? I re-gapped the sensor to .041”. Car starts & runs!
Here’s where it gets weird: AFR’s are way off. Rich by 3 points in some places in the map- but not all places! It's rich at idle and cruise, but up in the +100kpa reigon, it’s only rich by like .5AFR. Now my first thought was that the EGO sensor might have recovered from having fuel dumped on it, but it’s kinda out of wack now. So this morning, I recalibrated my EGO sensor. No change. It reacts fine, but it just reads really rich. My experience with bad EGO sensors is that they start getting really unpredictable. Also, when the car is dipping super rich (<10.5AFR), the car bogs hard like you would expect it to at that AFR, which makes me think it’s actually reading correctly. It’s also stalled twice when coming to a stop & transitioning to idle when the car is cold, and this is another situation where the AFR’s can swing a bit rich. I also have 2 symptoms which may be related to a misbehaving EGO controller, or might all be part of the same root-cause: 1) at temp, my idle KPA used to be ~29, and now it’s ~32. 2) my idle timing is hunting just a little bit- maybe max ~2* between 16-18 instead of being stable at 17. I don’t recall it wobbling like this before.
I’m trying to figure out WTF has changed that would affect AFR’s so much?
Timing is still correct per a timing light.
VVT target & actual are still following each other nicely on a log, although I haven’t done any recalibration of VVT settings in TS after swapping the actuator (would I need to?)
I ordered up another EGO sensor, so I can double check the sensor in a few days.
Did you by chance knock the vacuum line off the fuel pressure regulator? With no vacuum at the regulator you'd see more fuel pressure causing a rich condition. Happened to me a while back. Fredb
I haven't checked the woodruff key, but base timing is still spot on, which indicates to me that the crank hasn't walked. I checked the cams, and they appear to be timed correct.
Edit* no I'm thinking of this wrong. The timing wheel could walk relative to the crank. It's possible. I dread pulling the crank bolt now. The sway bar is in the way of getting my torque wrench on the crank, and the IC pipes are in the way of the sway bar mounting bolts.
You can see the timing marks on the pulley any time. You can see the timing mark on the crank sprocket with the crank pulley removed. You can only see the position of the crank sprocket relative to the crank if you pull the crank bolt.
I just installed a brand new ego sensor, so I'm going to see if that was the only culprit.
New ego sensor installed. No change. Still runs super rich.
Pulled the timing belt again. Took pics of timing for your viewing pleasure. Everything looks fine. Pulled the main crank bolt. Keyway looks fine.
Next I'll pull plugs and check compression. About the only thing I can think of at this point is maybe I bent a valve? I know this motor does actually interfere a little bit.
In the pictures, your timing belt looks wet. I assume you replaced the timing belt once you fixed the oil leaks? Otherwise, it might be stretched. You mentioned being surprised at how much additional slack you took up with the tensioner so . . . .
Beyond that, might be time to go autotune a bit? In the warmer months, the car will tend to run richer on the basemap due to lower air density. Temp sensor is supposed to correct for that, but isn't perfect.
I reused the belt. It has about 5k on it. I'm replacing the crank bolt and woodruff key this time around. I'll drop in a new belt too.
I'm not convinced the oil leak is fixed either. Still very wet in there.
Regarding autotuning, I'm suspicious of just tuning to fix VE. I JUST had it dyno tuned. Really the only thing I can tell that's different is that I followed the correct belt tensioning procedure, whereas the belt may have been kinda loose during dyno tuning.
There's really only four places oil leaks up there, assuming your oil pump and pan are sealed and you aren't missing the AC bolt. Cam x2, crank, and VC. They're all really easy to detect if you know where to leak.
Yeah, I know what you're saying. It's starting to look like it's leaking out of the top of the oil pump up near where the water pump abuts it. Hard to tell yet. It's pretty messy.
The only way I found to reliably find an oil leak was to clean the #$%^& out of everything and then drop in some dye. Oil has a way of really traveling. In my case, I've found oil leaks from cam seals on one fresh build and then from the little bleed screw not sealing well on another. In the first instance, the cam seals were installed correctly but came from a cheapo gasket set. I put in Mazda OEM and problem gone. Can't explain the difference though . . . looked the same to the naked eye.
Block has about 107k on it. I didn't pull & reseal the oil pump when I refreshed it at 102k, although I did reseal the pan. I just replaced the cam & crank seals with OE. I'll double check my AC bolt and the dipstick plug on the pump.
Is your engine at TDC in the picture showing the full camshafts? the lobes for cylinder 1 are pointed in the wrong direction (they should point slightly toward each other) Did you have the camgears off?
Edit: NEVERMIND - I've been looking at the top end of too many engines lately. What you have is correct for the Miata.
Got a new timing belt, crank bolt & woodruff key in. I noticed that the tensioner pulley pivot had come off the pivot pin at some point. Maybe that was what was giving me all my grief? Car started up ok. I'll try getting it up to temp & see what the AFR's look like this evening.
went for a drive & did some auto tuning. So far, I'm staying in the conservative part of the map. It's needed to pull a lot of fuel out of the map. This is the percentage reduction. Obviously, not all areas have been tuned.
Have you done a leakdown test with the new head? I could imagine leaky intake valves to cause going rich: the piston pushes air mixed with fuel out the intake valve during compression stroke. Then the injectors ad more fuel the next time around. Kind of like an involuntary Atkinson cycle.
07-11-2017, 09:24 PM
timing belt job causes rich AFR? What have I broken this time?
0
